The purpose of this project is to provide independent consultation to researchers within the NCI on problems related to biostatistical methodology and cancer control. Current projects include: -in collaboration with the Cancer Prevention Studies Branch, the American Association of Retired Persons (AARP) Observational Cohort Study has recently started; this project will construct a cohort of AARP members to investigate the relationship between current dietary intake and future incidence of cancer of the breast, colon/rectum, and prostate. -in collaboration with the Applied Research Branch, a newly developed food frequency questionnaire is being evaluated; a study to estimate the food frequency accuracies of a selected sample of questionnaire items began in 1995; an RFP is being developed to estimate the new questionnaire's measurement errors of selected nutrients and to compare its error to that of two other commonly used FFQs; robust regression methods to convert food frequencies into nutritional intakes based on the subject's age, gender and FFQ-reported portion size are being developed and evaluated. -in collaboration with the Cancer Prevention Studies Branch. analyses were conducted on the effects of micronutrient supplementation on immunology measures among subjects in the Linxian, China intervention study. -in collaboration with the Cancer Prevention Studies Branch data from the Breast Cancer Detection Demonstration Project Followup Study have been used to examine the relationship between relative weight (weight/height-squared) and breast cancer risk. -in collaboration with the Cancer Prevention Studies Branch data from the alpha-Tocopherol beta-Carotene Cancer Prevention Study conducted among Finnish male smokers was used to examine the relationship between alcohol consumption and the occurrence of colon cancer. -in collaboration with the Applied Research Branch, analyses of food frequency data from the 1987 National Health Interview Survey were undertaken. Patterns of food and fat intake of whites, blacks, and Hispanics were compared. -in consultation with the Applied Research Branch, DCPC, development of methods to evaluate the accuracy of the Cancer Information System services to the general public is continuing. -in collaboration with the Laboratory of Chemoprevention, DCE, we have analyzed results of experiments on combining 9-cis-retinoic acid with raloxifen against mammary tumors, on combining two different retinoids, 5531 and 4-HPR, with tamoxifen against prostate tumors, and on combining two different deltanoids, Ro 23-7553 and Ro 25-8808, with tamoxifen against mammary cancer. -in collaboration with the Prevention and Control Extramural Research Branch and the Applied Research Branch, the results of the 5-A-Day for Better Health Program was evaluated. Advice was given on the nature and conduct of a planned follow-up survey. Guidance for grant applicants in sample size estimation was also provided. -in collaboration with the Applied Research Branch, analyses of food frequency data from the 1987 National Health Interview Survey were undertaken. Patterns of food and dietary fat intake of whites, blacks, and Hispanics were compared -in collaboration with the Environmental Epidemiology Branch, DCE, the effects of long-term storage and multiple freeze-thaw cycles on measurement of serum ascorbic acid were assessed using samples from a cervical cancer case-control study.